Why most MSPs should not niche down to a single industry
Every MSP coach says growth means niching down to one industry. For a referral-driven local MSP, that shrinks the serviceable market and amputates the referral network. The real lever is right-fit clients, not a vertical.
By Stacey Tallitsch | June 18, 2026
You are 18 months past the point where the business stopped feeling fragile. Recurring revenue covers payroll, the help desk runs without you in the room, and you have maybe 40 clients who renew without a conversation. Then you sit in a peer group, or you finish a coaching call, or you read the same article for the third time this quarter, and the message is identical every time. Pick a vertical. Stop being a generalist managed service provider. Become the firm that does IT for dental practices, or law firms, or accounting offices. Rewrite the website, narrow the pitch, command a premium. The advice is delivered as settled fact. For a specific kind of MSP, it is correct. The problem is that most MSPs being told to niche down are not that kind.
The advice is real, and so is the math it ignores
Start with what the niche argument gets right, because it gets a lot right. Roughly 80% of the technology work an MSP does is identical across industries: patching, backup, endpoint security, the same stack regardless of who is sitting at the keyboard. The remaining 20% is where a vertical feels understood, and buyers choose on feeling. A focused pitch reaches more of the right people with less effort. Specialization reads as authority, and authority supports higher prices. None of that is wrong.
What the advice almost never does is count. Niching is a marketing-reach strategy, and it assumes reach is your constraint. For most MSPs under a few million in recurring revenue, reach is not the constraint. Geography is.
Here is the arithmetic the coaching deck skips. Your serviceable market is not "every dental practice in America." It is dental practices, in your metro, in the size band you can profitably serve, that are not already locked into a competitor's multi-year agreement. Per the SBA Office of Advocacy's 2024 small-business profiles for metropolitan areas, even the largest metros hold a finite, countable number of employer businesses in any single industry, and mid-size metros hold far fewer. Take a real second-tier metro. Filter to one vertical. Filter again to the 10-to-75-seat range you actually want. Subtract the ones already under contract. You can land on a serviceable market of 200 to 400 accounts. You need to win maybe 30 of them over several years, through a sales motion that is mostly referral.
That is the number nobody puts on the slide.
You are referral-driven, and niching narrows the room you refer inside
The average MSP supports around 122 clients, and most carry fewer than 100. That book was not built through paid funnels. Most MSPs, maybe the majority, are built almost entirely on word of mouth, and referrals convert at 3 to 5 times the rate of cold outreach. Your growth engine is the room: the local business groups, the accountant who sends you three offices a year, the client who tells the guy he golfs with. That room is mixed. It does not sort itself by industry.
When you niche down by industry, you do not just narrow your marketing. You narrow the surface area of your referral network to the slice of it that happens to touch your one vertical. The accountant who would have sent you a manufacturer, a nonprofit, and a medical office now sends you only the medical office, because you told everyone that is all you do. In a practice where the pipeline is built on relationships rather than reach, voluntarily shrinking the set of referrals you are eligible to receive is not focus. It is amputation dressed as strategy.
This is the same error I see when a smaller operator copies a larger competitor's playbook move for move. The niche-down advice was forged by MSPs and coaches operating at national scale, where a vertical can be reached through trade media, conference circuits, and a real marketing budget, and where 400 prospects is a rounding error because the prospect pool is the entire country. Import that strategy into a referral-driven local practice and you have imported the playbook's constraints without its conditions.
This is not a niche problem. It is a fit problem.
Here is the reframe that matters. The pain the niche advice promises to cure is real: generic positioning, scattered referrals, price compression, clients who churn. But the cause is almost never that you serve too many industries. The cause is that you serve too many kinds of client. Those are different variables, and the deck conflates them.
An MSP bleeding margin usually has a fit problem hiding under an industry costume. It has 5-seat clients on the same plan as 60-seat clients. It has break-fix holdovers paying hourly next to fully managed agreements. It has one customer with a compliance posture nobody scoped, and three who negotiated the contract down until the relationship loses money every month. None of that is fixed by deciding to do IT for law firms. All of it is fixed by defining a right-fit client by seat count, contract structure, security requirements, and willingness to be managed rather than rescued, and then disqualifying the accounts that do not match.
Right-fit discipline does what niching is supposed to do, without paying the geography tax. You get a repeatable pitch, because you are selling the same engagement shape every time. You get scalable delivery, because your stack and your runbooks assume one kind of client. You get premium pricing, because you stopped competing for the accounts that only ever bought on price. And you keep your entire referral network intact, because "I take well-run 20-to-60-seat companies that want IT handled, not patched" is a filter your accountant can repeat at a lunch, across every industry he serves.
Who should niche anyway
Be honest about the exceptions, because they are real and the advice exists for a reason. Niche down by industry when the 20% that feels unique actually is unique: a genuine compliance or workflow specialization, like healthcare, legal, or financial services, where the regulatory surface is deep enough that a generalist truly cannot compete. Niche down when you have outgrown your geography and need a way to reach prospects you will never meet at a chamber breakfast, because at that point a vertical is your only affordable path to national reach. Niche down when you already have the marketing budget and the content engine to own a category rather than just claim one.
If you are a sub-$3M MSP whose last 10 clients all arrived because someone knew someone, none of those conditions are true yet. The advice is not wrong. It is early. Adopting it now, before you have outgrown the room you sell inside, is the same category mistake as buying the wrong marketing service because it was the default pitch: right tool, wrong stage.
What to do before you narrow anything
Do not rewrite your website this quarter. Do this instead, and you can start today. Pull your last 20 clients into a spreadsheet. Tag each one by seat count, contract type, gross margin, and how they found you. Ignore industry entirely on the first pass. The pattern that separates your profitable, low-drama clients from your margin-killers will almost never be their industry. It will be their size, their contract, and whether they wanted to be managed. That pattern is your real ideal client. Write it as one sentence your best referral source could say out loud. Then count, honestly, how many accounts in your actual geography match it before you ever consider cutting the market down to a single industry. Narrowing is a decision you can only make safely after you have done the counting. Most MSPs niche first and count never.
— Stacey Tallitsch, Stronghold CMO
About the Author
Stacey Tallitsch is the President of Stronghold CMO, a Fractional AI CMO service operating under Talisman Capital, Inc. He is a 30-year tech veteran and the author of 21 books on systems thinking, operator-grade decision-making, and personal sovereignty, with more than 30,000 students across his Udemy course catalog.